Erection cycles during sleep are often viewed as one of the more unusual physiological events to occur during sleep, yet penile erections are a robust peripheral manifestation of rapid eye movement (REM) sleep in all healthy males. These REM-related erections were previously termed “nocturnal penile tumescence” (NPT), but are now more commonly referred to as “sleep-related erections” (SRE) to reflect the specific association with sleep. The history of SRE testing from its discovery to its clinical role in evaluating male erectile dysfunction is reviewed in this chapter. SRE testing was commonly performed in many sleep laboratories until a decade ago. Although urologists continue to utilize SRE testing using home screening devices, few sleep medicine specialists employ this technique today even though a clear role for formal in-laboratory SRE monitoring remains. Finally, the discovery of an animal model to elucidate some of the basic neural mechanisms of SREs is presented.

Prior to any description of SREs in the scientific literature, it had long been known that men commonly awaken out of sleep with an erection or even a nocturnal emission. Awakening with an erection has erroneously been believed a consequence of a full bladder or the need to void. This belief has persisted to this day in the lay public even though SREs have no known association with bladder fullness [1]. From a historical perspective, erections during sleep were commonly viewed by many, and particularly within the religious community, as representing “unhealthy” thoughts or dreams. As a result, efforts were undertaken in some circles to utilize devices designed to prevent erections during wakefulness or sleep. Such devices included “spermatorrhea rings” that comprised a flexible inner ring that was placed around the penis and attached to an inflexible outer ring containing metal teeth that were designed to inflict pain as the penis engorged during an erection when it came in contact with the teeth of the outer ring [1]. Numerous barbaric looking devices were utilized for this purpose, some with electrical switches involving circumferential rings triggered by expansion that would activate buzzers or bells to warn the user or parental guardian of sinful exploits transpiring under the covers. There are reports of adult patients who have been traumatized by such devices during their youth [1].

Although Halverson was the first to report visual observations of erections in infant males during sleep [2], the first systematic evaluation of erections during sleep appearing in the scientific literature was performed by Ohlmeyer et al. in 1944 [3], approximately 10 years prior to the discovery of REM sleep. These authors performed a series of penile erection recordings during sleep in healthy male volunteers using a simple “on–off” type switch that recorded the presence or absence of an erectile event. This recording device included a flexible inner ring placed around the shaft of the penis that was connected to an inflexible outer metal ring (but without the menacing teeth of the “spermatorrhea ring” described above). During an erection, the flexible inner ring came in contact with the outer ring, forming a closed electrical circuit that could be recorded during the night. Using this recording technique, Ohlmeyer and colleagues found that all normal healthy males in the study generally demonstrated three to four erection cycles lasting approximately 25 min each in duration and occurring every 85 min during sleep, similar to the cycling of REM sleep discovered by Aserinsky and Kleitman in 1953 [4].

Several authors, including Aserinsky, were aware of the erection data described by Ohlmeyer and colleagues a decade earlier and speculated that erections may be occurring during REM sleep given the similar cyclicity and duration of these two physiological phenomena during sleep. However, it was not until the mid-1960s that Fisher [5] and Karacan [6, 7] independently demonstrated for the first time that erections during sleep are a naturally occurring phenomenon of REM sleep in all healthy males. Early research also demonstrated that women exhibit similar genital activity during REM sleep, including clitoral erections [8], increased vaginal blood flow [9, 10], and increased intrauterine pressures [11] occurring in a cyclical manner during REM sleep as observed in men.

Fisher [5] and Karacan [6, 7] had both utilized an erection recording device involving mercury loop strain gauges. This device involves a mercury-filled silastic tubing that carries a small electrical current passing through the mercury in the loop. During an erection, the silastic tubing stretches as it accommodates the increase in penile circumference, and the column of mercury within the tubing becomes thinner, thus increasing the electrical resistance through the mercury-filled loop. Within a limited range of circumference changes, the gauge is relatively linear in that any given change in circumference relates to a specific change in resistance that may be monitored during sleep with polysomnography [12].

The strain gauge technology became the standard recording method with polysomnography from the 1960s to the present day. Two gauges are typically used for each patient, one placed at the base of the penis and the other just caudal to the glans in the coronal sulcus. Although the strain gauge technology allows for a precise determination of circumference changes during sleep, circumference alone is inadequate for an evaluation of erectile functioning. Some individuals may exhibit a small increase in penile circumference yet demonstrate a maximally rigid erection, whereas others may have a marked increase in circumference with minimal rigidity.

The buckling force device was thus developed to formally evaluate penile rigidity in the sleep laboratory. The patient is awakened during a maximal circumference increase either during or at the end of a REM sleep episode. The technician quickly stabilizes the base of the penis between the thumb and index finger and then presses the buckling force device against the head of the glans along the longitudinal axis of the penis. The pressure at which the penile shaft first bends or buckles is defined as the buckling force pressure. A penis that buckles at less than 500 g of force is generally considered inadequate for vaginal penetration [13]. At least two buckling force measurements are generally performed on every patient.

Other techniques were attempted to monitor erections during sleep in the mid-1960s prior to the strain gauge method becoming the standard of care. Fisher and colleagues in their original publication [5] had also utilized a polyvinyl tube with the size and shape of a doughnut that was filled with water and placed around the base of the penis. During an erection, the increase in circumference of the penis would displace the water in the doughnut-shaped tube and cause a rise in a water-filled column next to the subject’s bed. This device, however, was bulky and cumbersome. Finally, temperature monitoring of penis had also been utilized and was found to demonstrate an increase in the temperature of the penis during erection associated with the increase in blood flow during the erection [5], but this technique could not quantify the quality of erectile events during sleep. The ability of the mercury-filled strain gauge to accurately quantify changes in penile circumference and its minimal intrusiveness made this technique the preferred method of monitoring erections during sleep for the several decades that followed.

Karacan [14] was the first to suggest that monitoring erections during sleep may provide a clinical tool to differentiate psychogenic from organic erectile dysfunction. Fisher [15, 16] and others also appreciated the clinical value of involuntary erection cycles during sleep, leading to the new development of combining polysomnography with SRE monitoring to evaluate male sexual function. This new technology of SRE monitoring came at a time when erectile dysfunction was poorly understood and provided a new avenue for research opportunities to study sexual function. Masters and Johnson [13] had just published their seminal work suggesting that most male impotence was psychological in origin. However, new interventions were being developed to treat erectile dysfunction, including surgical penile implants, and a greater appreciation of organic or physiological causes of erectile failure were only beginning to be appreciated.

In the mid-1970s and long before the advent of oral medications to treat erectile dysfunction, two primary treatment options were available to men with impotence. Psychotherapy was offered if the impotence was believed to be psychological in origin, whereas a surgical penile prosthesis was considered for men whose erectile failure was thought to be organic or physiological in nature. SRE monitoring with polysomnography provided a new diagnostic technique to guide clinicians to the appropriate treatment option. For example, a patient who complained of difficulty initiating or maintaining an erection but was found to have normally occurring erection cycles during sleep was deemed to have a psychogenic cause of the erectile failure. On the other hand, a patient found to have a decrease or absence in erectile activity during sleep in the presence of adequate REM sleep would suggest that such an individual has an organic etiology of the erectile dysfunction.

SRE monitoring for the clinical evaluation of erectile capability lead to the need of establishing normative data. Karacan and coworkers [7, 17–21], and to a lesser extent Fisher [22], studied penile erections during sleep in normal boys and men of all age groups, demonstrating that erections occur in a tight temporal association with REM sleep in all healthy males tested from infancy to old age. REM-related erections were demonstrated to begin several minutes prior to the onset of REM sleep, and to last several minutes following the termination of REM when detumescence would occur. Later work replicated these findings [23–25]. Although some minor decreases in erection time during sleep are found in men as they age from adolescence to later years, the occurrence of erections as part of the normal physiology of REM sleep is a robust and predictable phenomenon in healthy, potent males, thus establishing SRE monitoring as a valid tool to evaluate erectile function.

Further work established that erections during sleep are generally unaffected by many psychological influences, thus strengthening the clinical value of SRE monitoring as a technique to differentiate psychogenic from organic erectile dysfunction. For example, neither presleep sexual activity nor abstinence appears to alter erections during REM sleep [6, 26–28]. Although the total tumescence time may be decreased in milder forms of depression or anxiety, penile erections persist during REM as a robust phenomenon in these patients with the total tumescence time still exceeding the total REM sleep time [1]. On the other hand, SREs may be adversely affected in some patients with major depressive disorders [29] . It remains to be determined to what extent fragmentation of REM sleep, the proerectile stimulus, in patients with major depression may be the cause of decreased erectile activity during sleep in this select group of patients.

Over the past several decades, a rare disorder has been described in some patients who frequently awaken with painful erections during sleep [30], yet describe no pain associated with erections during sexual intercourse or masturbation [31–34]. Although potential roles for neurovascular compression of the hypothalamus [35] or beta-adrenergic hypersensitivity [32] have been proposed as possible etiological factors, this disorder is poorly understood. Patients typically awaken several times during the night out of REM sleep with a painful erection over a period of many years. Such patients tend to have fragmented REM sleep, which is thought to be secondary to the pain associated with the erection. Patients often have a history of anxiety, but no other clear pathology has been identified. Unfortunately, there are few treatment options for such patients other than some transient relief reported with the beta-blocker propranolol and some sustained benefit with clozapine [32, 36, 37].

Many comorbid medical conditions or cardiovascular risks have been found to adversely affect SREs, and many of these same conditions are also known to adversely impact normal, or waking, erectile functioning. For example, SREs have been demonstrated to be decreased in patients with diabetes [38, 39], hypertension [40], smoking [41], alcoholism [42, 43], and intoxication [44]. Men with hypertension and complaints of erectile dysfunction demonstrate significantly less total tumescence time during sleep compared to hypertensive men without erectile complaints. Although many of these comorbid conditions that increase cardiovascular risks are hypothesized to adversely impact erectile capability at the peripheral or end organ level, the impact of such disease states on the central nervous system (CNS) control of erections during wakefulness or sleep has not been adequately examined. Indeed, some evidence has been presented to suggest that diabetes may decrease SREs long before peripheral neurovascular abnormalities are apparent, suggesting that at least some comorbid conditions may impact the CNS control of erections during sleep [45].

Considerable data have been published on SREs with respect to either hypogonadism or the manipulation of androgen levels in males with both low and normal testosterone levels. Although these data show that SREs are adversely affected by low androgen levels, there has been some controversy in the literature over the years as to what extent SREs are “androgen dependent.” Testosterone is released during sleep in normal adult males and tends to peak at a time during REM sleep when erections occur [46]. Moreover, testosterone administration has been shown to increase SREs in hypogonadal men [47]. On the other hand, discontinuing testosterone replacement in hypogonadal patients decreases the total tumescence time during sleep, the maximal penile circumference, and the number of tumescence episodes during sleep [48]. These and other data suggest that androgens play an important role in augmenting erections during sleep in adult males [49–52], leading some to suggest that SREs are “androgen dependent” [53]. However, SREs are prominent in infant and prepubescent males [21] in the face of undetectable testosterone levels. Indeed, children exhibit more tumescent time during sleep than adult males secondary to the increase in total REM sleep time in children [21]. Finally, although SREs may be decreased in low androgen states, SREs in hypogonadal males persist and often remain within the normal range [48]. These later data have led to the current view that SREs may be better termed “androgen sensitive” [54, 55] instead of the previously used terminology of “androgen dependent.”

One of the major developments in the management of erectile dysfunction has been the growing understanding that obstructive sleep apnea (OSA) may adversely impact erectile functioning during wakefulness. Guilleminault and coworkers reported on an early case series in 1977 that men with OSA had a high prevalence of erectile complaints, and that some men reported improvements in erectile capability following treatment with nasal continuous positive airway pressure (CPAP) [56]. Four years later Schmidt and Wise reported on a series of 15 consecutive patients with impotence presenting to a urology clinic and found that those with organic erectile dysfunction ( n = 7) had a significantly higher prevalence of OSA compared to the group with pscychogenic impotence [57], a finding that was confirmed by Pressman [58] and Hirshkowitz [59]. Karacan and Karatas later reported that one third of patients with OSA demonstrate improved SREs following nasal CPAP therapy [60].

In spite of these early data, it was not until the mid-2000s that the concept of OSA as a potential independent risk factor for the development of erectile dysfunction began to be more seriously entertained. Numerous studies now suggest that sleep apnea patients are at a significantly higher risk for erectile dysfunction and that nasal CPAP may improve erectile capability [61–65], confirming and extending the earlier studies. Finally, intermittent hypoxia during sleep decreases waking erectile functioning in rats [66] and mice [67] using experimental designs that manipulate blood oxygen saturation. Given that OSA has been well documented to be a risk factor for several cardiovascular diseases , as well as a potential independent risk factor for erectile dysfunction during wakefulness, many urology clinics are now screening for OSA in patients presenting with erectile dysfunction as part of a treatment regimen addressing potential modifiable risk factors for impotence [68].

Several techniques have been utilized over the years to evaluate erectile functioning during sleep without involving formal polysomnography. As early as 1980, it was suggested that the cost of erection monitoring during sleep could be reduced from hundreds of dollars in the sleep laboratory to as little as “30 cents” for three nights of home testing using the so-called stamp-ring method [69]. This technique involved using four or five US postage stamps wrapped around the flaccid penile shaft and wetting the overlapping stamp to seal the “stamp ring” prior to bedtime [69]. If the stamp ring was broken the next morning upon awakening, it was presumed that SREs were normal and erectile functioning was intact [70]. However, the validity of this technique was put into question since a breaking of the stamp ring along the perforations between the stamps was found to be an unreliable indicator of normal penile rigidity [71].

These concerns led to the “snap-gauge band,” a penile ring that attempted to utilize multiple breakable bands within the ring with varying levels of tension required to break each band in an attempt to quantify the degree of circumferential rigidity obtained during the night of sleep [72–75]. Eventually, similar concerns of reliability were raised with this snap-gauge technique in that some men may have marked increases in penile circumference during an erection, but with minimal rigidity required for vaginal penetration [76]. A formal comparison of the snap-gauge technique with in-laboratory polysomnography by Allen and Bender [77] also revealed the surprising finding than many men failed to break any of the snap-gauge bands even while demonstrating normal SREs and adequate penile rigidity as measured with the bucking force device. Formal polysomnography was continued to be viewed as the gold standard technique for monitoring SREs [78].

A more advanced screening device was introduced in 1986 called the RigiScan (Endocare, Irvine, CA). Still in use today, it is designed to record erections during sleep at home without monitoring sleep–wake stages [79]. The introduction of the Rigiscan home screening device was in part responsible for a move away from in-laboratory polysomnography as a means of evaluating erectile function during sleep. This recording system utilizes two loop devices placed around the penile shaft, one at the base and the other near the glans, each containing a moveable wire. The loop device has the advantage of being able to simultaneously monitor penile circumference and radial or circumferential rigidity. Increases in circumference of the penis during an erection are monitored via the wire in the loop. In addition, a squeezing force of 283.5 g is applied to the penile shaft at regular intervals during the night by pulling on the wire within the loop. The radial rigidity is measured in arbitrary units so that a 100 % radial rigidity corresponds to no measurable circumferential displacement from the 283.5-g force. For every 0.5-mm loop shortening that is detected, circumferential rigidity decreases by 2.3 %.

The RigiScan has the advantage of monitoring rigidity many times throughout the night, unlike the buckling force that provides only two to three isolated measurements of rigidity. However, circumferential rigidity as measured by the RigiScan device has been shown to be inferior to the buckling force method as a predictor of axial rigidity, which is the rigidity required for vaginal penetration [80]. Moreover, the RigiScan device does not involve the important contribution of patient and technician visual observations of the erection and/or discrepancies in the quality of the erectile event. During buckling force measurements in the sleep lab, both the technician and the patient rate the quality of the erection. Not only may physical abnormalities, such as Peyronie’s disease be observed during the nocturnal erection assessment, but also some patients with psychogenic erectile dysfunction tend to underestimate the true capability of erectile function as seen by a discrepancy between the technician and patient rating of the erection.

Urologists have continued to utilize SRE testing even though very few sleep medicine specialists provide this service today. The advent of home screening devices, as well as the introduction of phosphodiesterase inhibitors such as Viagra, has reduced the perceived need for in-laboratory evaluations. Indeed, most SRE monitoring at this time is limited to the RigiScan home screening device [81]. Those who advocate home screening have argued without evidence that monitoring erections at home is more “natural” than in the laboratory, or that in-laboratory testing is not cost effective as a screening tool [81]. However, Schmidt and Schmidt [82] reviewed data demonstrating that a formal in-laboratory SRE evaluation continues to be the “Gold Standard” to evaluate erectile function and may be cost-effective if limited to specific clinical indications such as nonresponders to commonly prescribed medications for erectile dysfunction or those seeking a potential surgical prosthesis as a treatment option, a procedure that would destroy any remaining natural erectile function. Moreover, in-laboratory SRE testing would have particular relevance for medicolegal cases evaluating erectile capability for individuals claiming erectile dysfunction following pelvic injury or when impotence is used as a defense for a male accused of sexual assault [82]. Finally, in-laboratory polysomnography would also identify other potential risk factors contributing to erectile dysfunction, such as OSA. These guidelines for in-laboratory SRE monitoring have been published in detail elsewhere [1, 82].